COX-2 rs689466, rs5275, and rs20417 polymorphisms and …Wei-Dong Leng1†, Xiu-Jie Wen2†, Joey S. W. Kwong3†, Wei Huang4, Jian-Gang Chen5 and Xian-Tao Zeng1,5* Abstract Background:

RESEARCH ARTICLE Open Access

COX-2 rs689466, rs5275, and rs20417polymorphisms and risk of head and necksquamous cell carcinoma: a meta-analysisof adjusted and unadjusted dataWei-Dong Leng1†, Xiu-Jie Wen2†, Joey S. W. Kwong3†, Wei Huang4, Jian-Gang Chen5 and Xian-Tao Zeng1,5*

Abstract

Background: Numerous case–control studies have been performed to investigate the association between threecyclooxygenase-2 (COX-2) polymorphisms (rs20417 (−765G > C), rs689466 (−1195G > A), and rs5275 (8473 T > C))and the risk of head and neck squamous cell carcinoma (HNSCC). However, the results were inconsistent. Therefore,we conducted this meta-analysis to investigate the association.

Methods: We searched in PubMed, Embase, and Web of Science up to January 20, 2015 (last updated on May 12,2016). Two independent reviewers extracted the data. Odds ratios (ORs) with their 95 % confidence intervals (CIs)were used to assess the association. All statistical analyses were performed using the Review Manager (RevMan)5.2 software.

Results: Finally 8 case–control studies were included in this meta-analysis. For unadjusted data, anassociation with increased risk was observed in three genetic models in COX-2 rs689466 polymorphism;however, COX-2 rs5275 and rs20417 polymorphisms were not related to HNSCC risk in this study. Thepooled results from adjusted data all revealed non-significant association between these three polymorphismsand risk of HNSCC. We also found a similar result in the subgroup analyses, based on both unadjusted data andadjusted data.

Conclusion: Current results suggest that COX-2 rs689466, rs5275, and rs20417 polymorphisms are not associatedwith HNSCC. Further large and well-designed studies are necessary to validate this association.

Keywords: COX-2 rs689466, COX-2 rs5275, COX-2 rs20417, Polymorphism, Head and neck squamous cellcarcinoma, Meta-analysis

BackgroundHead and neck squamous cell carcinoma (HNSCC) is1 of the disease burdens worldwide affecting eating,breathing, and appearance. Besides environmentalrisk factors, such as tooth loss [1], alcohol consump-tion [2], periodontal diseases [3], smoking [4], tooth

brushing [5], and human papillomavirus (HPV) [6],genetic factors [7, 8] also play an significant role inthe onset and development of HNSCC. Many poly-morphisms have been identified associated with riskof HNSCC by meta-analyses, such as the hOGG1Ser326Cys polymorphism [9], XRCC1 Arg194Trppolymorphism [10], ERCC2 rs1799793 and rs13181polymorphisms [11]; however, some polymorphismsincluding XPD Asp312Asn polymorphism [12], TP53codon 72 polymorphism [7], and VEGF gene poly-morphisms [13] are not associated with HNSCC risk.Particularly within the same gene, theXRCC1genefor example, XRCC1 Arg194Trp polymorphism was

* Correspondence: [email protected]†Equal contributors1Department of Stomatology, Taihe Hospital, Institute of Oral andMaxillofacial Surgery, Hubei University of Medicine, Shiyan 442000, China5Department of Stomatology, Center for Evidence-Based and TranslationalMedicine, Zhongnan Hospital of Wuhan University, 169 Donghu Road,Wuhan 430071, ChinaFull list of author information is available at the end of the article

© 2016 Leng et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Leng et al. BMC Cancer (2016) 16:457 DOI 10.1186/s12885-016-2535-3

http://crossmark.crossref.org/dialog/?doi=10.1186/s12885-016-2535-3&domain=pdfmailto:[email protected]://creativecommons.org/licenses/by/4.0/http://creativecommons.org/publicdomain/zero/1.0/

associated with increased risk while Arg399Gln andArg280His polymorphisms were not [10].The human cyclooxygenase-2 (COX-2), the key en-

zyme in the conversion of arachidonic acid to pros-tatglandins, is located at chromosome 1q25.2-q25.3and rs20417 (−765G > C), rs689466 (−1195G > A), andrs5275 (8473 T > C) are the three commonly investi-gated polymorphisms in the COX-2 gene [14, 15].Now the association between COX-2 gene polymor-phisms and risk of many cancers, such as hepatocel-lular carcinoma [16], colorectal cancer [17], breastcancer [18], prostate cancer [19], gastric cancer [20]were investigated by meta-analyses. COX-2 has beenconfirmed very low or no expression in normal hu-man oral tissues, otherwise it was elevated in oralprecancerous lesions and over-expressed in oral squa-mous cell carcinoma (OSCC) [21]. The elevated ex-pression of COX-2 was presented to be correlatedwith malignant transformation, advancing clinicalstage, and disease progression [22].There are also many published studies that ex-

plored the association between COX-2 rs689466,rs5275, and rs20417 polymorphisms and risk of HNSCC.

Unfortunately, the results of published studies were incon-sistent and using a meta-analytic method to pool these re-sults for obtaining a more precise result [23] is necessary.In this meta-analysis, we extracted and combined crudedata and adjusted data.

MethodsWe reported this meta-analysis according to the PreferredReporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [24] and ethical approval isnot necessary.

Eligibility criteriaCohort studies or case–control studies evaluating therisk of HNSCC in relation to COX-2 rs689466,rs5275, and/or rs20417 polymorphisms were consid-ered for eligibility if they also met the following cri-teria: (1) the cancer was HNSCC, oral squamous cellcarcinoma (OSCC), or laryngeal squamous cell carcin-oma (LSCC) confirmed using microscopic examin-ation; (2) the frequency of genotype distribution,adjusted odds ratios (ORs) and their 95 % confidenceintervals (CIs), or the data that can calculate them

Fig. 1 Study selection flowchart

Leng et al. BMC Cancer (2016) 16:457 Page 2 of 12

Table 1 Characteristics and unadjusted data of included studies

Study Country (Ethnicity) Form ofdisease

Cases/Control HWE Smokingstatus

GenotypingmethodsSample size Genotype distribution

rs689466 (−1195G > A) GG GA AA

Chiang 2008 China (Asian) OSCC 368/441 80/114 187/235 101/92 Yes Mixed PCR-RFLP

Peters 2009 Netherlands (Caucasian) HNSCC 431/438 275/260 134/163 22/15 Yes Mixed PCR

Mittal 2010 India (Asian) OSCC 193/137 3/5 57/32 133/100 Yes Smokers PCR-RFLP

Chang 2013 China (Asian) HNSCC 313/295 93/90 146/148 74/57 Yes Mixed Taqman

Niu 2014 China (Asian) HNSCC 259/1035 61/222 126/542 72/271 Yes Mixed Taqman

OSCC 140/1035 44/222 80/542 25/271 Yes Mixed Taqman

LSCC 90/1035 17/222 46/542 27/271 Yes Mixed Taqman

rs5275 (8473 T > C) TT TC CC

Campa 2007 European (multicenter) HNSCC 553/711 252/313 237/321 44/77 Yes Mixed TaqMan

OSCC 252/711 113/313 117/321 22/77 Yes Mixed TaqMan

LSCC 281/711 139/313 120/321 22/77 Yes Mixed TaqMan

Mittal 2010 India (Asian) OSCC 135/59 74/24 53/34 8/1 No Smokers PCR-RFLP

Chang 2013 China (Asian) HNSCC 313/295 209/199 89/86 15/10 Yes Mixed Taqman

Niu 2014 China (Asian) HNSCC 258/1032 177/691 72/316 9/25 Yes Mixed Taqman

OSCC 168/1032 118/691 45/316 5/25 Yes Mixed Taqman

LSCC 90/1032 59/691 27/316 4/25 Yes Mixed Taqman

rs20417 (−765G > C) GG GC CC

Lin 2008 China (Asian) OSCC 297/280 193/107 104/173 0/0 Yes Mixed PCR–RFLP

Chiang 2008 China (Asian) OSCC 178/205 136/166 42/39 0/0 Yes Mixed PCR–RFLP

Peters 2009 Netherlands (Caucasian) HNSCC 428/433 321/321 99/99 8/13 Yes Mixed PCR

Mittal 2010 India (Asian) OSCC 176/96 92/41 78/49 6/6 Yes Smokers PCR–RFLP

Lakshmi 2012 India (Asian) OSCC 150/150 110/142 28/6 12/2 No Mixed PCR–RFLP

OSCC oral squamous cell carcinoma; HNSCC, head and neck squamous cell carcinoma; LSCC laryngeal squamous cell carcinoma; HWE Hardy–Weinberg Equilibrium

Lenget

al.BMCCancer

(2016) 16:457 Page

3of

12

Table 2 Adjustment and adjusted data of included studies

Study Form of disease Reference OR (95 % CI) Adjustment

rs689466 (−1195G > A)

Peters 2009 HNSCC GG: 1.00 GA: 0.79 (0.58–1.07); age (continuous), sex, smoking (continuous, 5 levels),and alcohol consumption (continuous, 3 levels)

AA: 1.24 (0.60–2.56)

Mittal 2010 OSCC GG: 1.00 GA: 3.07 (0.66–13.24); age, gender

AA: 2.22 (0.52–9.50)

G: 1.00 A: 1.03 (0.60–1.42)

Chang 2013 HNSCC GG: 1.00 GA: 0.86 (0.56–1.32); sex, age, education, cigarette smoking (pack-year categories),betel quid chewing (pack-year categories), and alcoholdrinking (frequency)AA: 1.23 (0.72–2.09);

GA + AA: 0.96 (0.64–1.43);

G: 1.00 A: 1.08 (0.83–1.40)

Niu 2014 HNSCC GG: 1.00 GA:0.85 (0.60–1.21) age, sex, smoking status, and drinking status

AA: 1.01 (0.69–1.50)

GA + AA: 0.91 (0.65–1.26)

OSCC GG: 1.00 GA: 0.74 (0.49–1.11)

AA: 0.87 (0.55–1.39)

GA + AA:0.78 (0.53–1.14)

LSCC GG: 1.00 GA:1.16 (0.65–2.09)

AA:1.43 (0.75–2.75)

GA + AA:1.23 (0.71–2.15)

rs5275 (8473 T > C)

Campa 2007 HNSCC TT: 1.00 TC: 1.03 (0.82–1.28); age, sex, center, tobacco consumption (packyears),and years of alcohol consumption

CC: 0.75 (0.51–1.10);

TC + CC: 0.97 (0.78–1.20)

OPSCC TT: 1.00 TC: 1.16 (0.86–1.58);

CC: 0.91 (0.53–1.54);

TC + CC: 1.11 (0.83–1.49)

LSCC TT: 1.00 TC: 0.88 (0.63–1.22);

CC: 0.60 (0.34–1.05);

TC + CC: 0.82 (0.60–1.12)

Mittal 2010 OSCC TT: 1.00 TC: 0.27 (0.03–2.26); age, gender

CC: 0.28 (0.03–2.33)

T: 1.00 C: 0.88 (0.55–1.40)

Chang 2013 HNSCC TT: 1.00 TC: 1.04 (0.69–1.56); sex, age, education, cigarette smoking (pack-yearcategories), betel quid chewing (pack-year categories),and alcohol drinking (frequency)CC: 1.89 (0.74–4.82);

TC + CC: 1.12 (0.75–1.65)

Niu 2014 HNSCC TT: 1.00 TC: 0.90 (0.66–1.22); age, sex, smoking status, and drinking status

CC: 1.48 (0.68–3.25);

TC + CC: 0.94 (0.70–1.26)

OSCC TT: 1.00 TC: 0.86 (0.58–1.26);

CC: 1.03 (0.36–2.97);

TC + CC: 0.87 (0.60–1.27)

LSCC TT: 1.00 TC: 1.02 (0.63–1.64);

CC: 1.62 (0.54–4.88);

TC + CC: 1.07 (0.67–1.69)


were reported; (3) full-text were obtainable; (4) if 2 ormore studies covered the same population, we in-cluded the study that contained most comprehensiveinformation; (5) the published language is English orChinese.

Search strategyWe searched PubMed, Embase, and Web of Science upto January 20, 2015 (last updated on May 12, 2016)using the following search terms: head and neck, oral,oral cavity, pharyngeal, oropharynx, laryngeal, laryngo-pharyngeal, mouth, tongue, carcinoma, cancer, tumour,neoplasm, cyclooxygenase-2, COX-2, PTGs2, poly-morphism, mutation, variant, and variation. We alsoscreened reference lists of recent reviews, eligible stud-ies, and published meta-analyses on related topics foradditional eligible studies.

Data extractionThe following data were extracted from all eligiblestudies by 2 authors independently and disagreements(κ = 0.96) were resolved by discussion: last name ofthe first author; publication year; country and ethni-city; genotyping method; source of control, numberand genotyping distribution of cases and controls;adjusted OR and its 95 % CI; adjusted variables; and

Hardy–Weinberg Equilibrium (HWE) for controls[25]. The meta-analysis reviewers were blind to thestudy author and institution of the studies undergoingreview.

Statistical analysisThe heterogeneity was assessed first using the CochraneQ and I2 statistic [26]. The heterogeneity was consideredacceptable if both p > 0.1 and I2 < 40 % and used thefixed effect model, otherwise the random effect modelwas used. For crude data, we used OR and its 95 %confidence interval (CI) to quantify the strength ofassociation using the allele comparison, homozygotecomparison, heterozygote comparison, dominant model,and recessive model genetic models. For adjusteddata, we directly combined the relevant ORs and their95 % CIs according to reported genetic models. Weperformed subgroup analyses based on ethnicity, siteof cancer, and HWE status for controls. The sensitiv-ity analysis was performed by switching the effectmodel. Publication bias was assessed by funnel plotsif the number of included studies was more than 9.All statistical analyses were performed using ReviewManager (RevMan) software (version 5.2 for Windows;Copenhagen: The Nordic Cochrane Centre, The CochraneCollaboration).

Table 2 Adjustment and adjusted data of included studies (Continued)

rs20417 (−765G > C)

Lin 2008 OSCC GG: 1.00 GC + CC: 0.22 (0.12–0.39) age, gender, ethnicity, educational level, and habits of betelquid chewing, cigarette smoking, and alcohol drinking

Peters 2009 HNSCC GG: 1.00 GC: 0.99 (0.71–1.40); age (continuous), sex, smoking (continuous, 5 levels), andalcohol consumption (continuous, 3 levels)

CC: 0.59 (0.23–1.49)

Mittal 2010 OSCC GG: 1.00 GC: 0.71 (0.42–1.18); age, gender

CC: 0.44 (0.13–1.46)

G: 1.00 C: 0.73 (0.50–1.08)

OSCC oral squamous cell carcinoma; HNSCC head and neck squamous cell carcinoma; LSCC laryngeal squamous cell carcinoma; OR odds ratio; CI confidence interval

Fig. 2 Forest plot for A vs. G model of crude data of rs689466 polymorphism


Table 3 Overall and subgroups meta-analysis of COX-2 rs689466 polymorphism and HNSCC risk

Overall and subgroups No. OR (95 % CI) Heterogeneity (I2%/p)

A vs. G (unadjusted and adjusted)

Overall (unadjusted) 5 1.08 (0.97–1.09) 7 %/0.37

Overall (adjusted) 2 1.07 (0.84–1.36) 0 %/0.88

Asians (unadjusted) 4 1.12 (1.00–1.25) 0 %/0.56

Asians (adjusted) 2 1.07 (0.84–1.36) 0 %/0.88

Caucasian (unadjusted) 1 0.92 (0.73–1.16) NA

OSCC (unadjusted) 3 1.01 (0.87–1.16) 80 %/0.008

OSCC (adjusted) 1 1.03 (0.60–1.42) NA

LSCC (unadjusted) 1 0.96 (0.72–1.32) NA

AA vs. GG (unadjusted)

Overall 5 1.26 (1.01–1.57) 0 %/0.46

Asians 4 1.25 (0.99–1.57) 14 %/0.32

Caucasian 1 1.39 (0.70–2.73) NA

OSCC 3 1.07 (0.40–2.86) 86 %/

ResultsStudy identification and characteristicsWe yielded 408 papers initially and 8 case–control stud-ies [27–34] were included finally, Fig. 1 showed the pro-gress of study selection. Of them, 5 case–control studiesinvolving 1564 cases and 2346 controls focused onCOX-2 rs689466 polymorphism [28, 30, 31, 33, 34], 4studies involving 1259 cases and 2097 controls onCOX-2 rs5275 polymorphism [27, 31, 33, 34], and 5studies involving 1229 cases and 1164 controls onCOX-2 rs20417 polymorphism [28–32]. One studydid not satisfy the HWE for COX-2 rs5275 poly-morphism [31] 1 for COX-2 rs20417 polymorphism[32]. The main characteristics are shown in Table 1and Table 2.

COX-2 rs689466 polymorphism and HNSCC riskThe pooled results from crude data indicated therewas a significant increased risk of association betweenCOX-2 rs689466 polymorphism and HNSCC risk inAA vs. GG, AA vs. GA, and AA vs. GG + GA geneticmodels while no association in A vs. G (Fig. 2) andAA + GA vs. GG genetic models. Subgroup analysesstratified by ethnicity and cancer site all revealednegative results. The results of adjusted data showedno association between COX-2 rs689466 polymorph-ism and HNSCC risk in overall population and sub-group analyses. The sensitivity analysis showed theresults without substantive change. Table 3 showedthe results of all analyses.

COX-2 rs5275 polymorphism and HNSCC riskThe pooled results of crude and adjusted data all showednonsignificant association between COX-2 rs5275 poly-morphism and HNSCC risk in overall population, Fig. 3showed the result of C vs. T model of crude data. Theresults of subgroup analyses all revealed negative associ-ation. The sensitivity analysis showed the results withoutsubstantive change. Table 4 showed the results of allanalyses.

COX-2 rs20417 polymorphism and HNSCC riskTable 5 presented the results of COX-2 rs20417 poly-morphism and HNSCC risk. All results from unadjusteddata and adjusted data presented nonsignificant associ-ation, either in overall or subgroups population; Fig. 4showed the result of C vs. G model of crude data. Thesensitivity analysis showed the results without substan-tive change.

Publication biasDue to the limited number of included studies, we didnot conduct publication bias analysis.

DiscussionThe rs20417, rs689466, and rs5275 polymorphismsare the three commonly investigated polymorphismsin the COX-2 gene [14, 15]. In 2007, Campa D et al.conducted a case–control study including 533 casesand 1066 controls which indicated no significant asso-ciation between COX-2 rs5275 polymorphism andHNSCC risk [27]. Then Chiang SL et al., in 2008,showed that COX-2 rs20417 polymorphism was notassociated with OSCC risk but COX-2 rs689466 wasassociated with increased risk of OSCC [28]. However,another study obtained this increased risk betweenCOX-2 rs20417 polymorphism and OSCC [29]. Simi-larly, published studies on these three polymorphismsrevealed inconsistent results. This meta-analysis basedon the crude data indicated there might be an associ-ation with increased risk of HNSCC in COX-2rs689466 polymorphism, but identified negative asso-ciation between COX-2 rs5275 and COX-2 rs20417polymorphisms and HNSCC risk. However, the com-bined results of adjusted data all yielded nonsignifi-cant associations between these three polymorphismsand HNSCC risk. The subgroup analyses according toethnicity and sites of HNSCC confirm this negativeassociation.This meta-analysis is the first study to investigate these

three polymorphisms and risk of HNSCC. Unlike the

Fig. 3 Forest plot for C vs. T model of crude data of rs5275 polymorphism




C vs. T (unadjusted and adjusted)



HWE (Yes–unadjusted) 3 0.94 (0.82–1.06) 0 %/0.45

HWE (No–unadjusted) 1 0.69 (0.42–1.11) NA


Asians (adjusted) 2 1.06 (0.81–1.40) 0 %/0.33


OSCC (unadjusted) 3 0.90 (0.76–1.06) 0 %/0.52

OSCC (adjusted) 1 0.88 (0.55–1.40) NA

LSCC (unadjusted) 2 0.88 (0.73–1.06) 47 %/0.17

CC vs. TT (unadjusted and adjusted)






Asians (adjusted) 3 1.45 (0.81–2.59) 16 %/0.30


Caucasian (adjusted) 1 0.75 (0.51–1.10) NA

OSCC (unadjusted) 3 0.92 (0.59–1.43) 0 %/0.48

OSCC (adjusted) 3 0.89 (0.56–1.40) 0 %/0.57

LSCC (unadjusted) 2 0.98 (0.35–2.75) 67 %/0.08

LSCC (adjusted) 2 0.88 (0.34–2.26) 60 %/0.12

CC vs. CT (unadjusted and adjusted)






Asians (adjusted) 3 0.93 (0.73–1.19) 0 %/0.49



OSCC (unadjusted) 3 0.99 (0.64–1.53) 44 %/0.17

OSCC (adjusted) 3 1.02 (0.80–1.30) 28 %/0.25

LSCC (unadjusted) 2 0.87 (0.54–1.40) 50 %/0.16

LSCC (adjusted) 2 0.92 (0.70–1.21) 0 %/0.62

CC vs. CT + TT (unadjusted)

Overall 4 0.96 (0.70–1.31) 43 %/0.15

HWE (Yes) 3 0.91 (0.66–1.25) 46 %/0.16

HWE (No) 1 3.65 (0.45–29.89) NA

Asians 3 1.58 (0.93–2.71) 0 %/0.70

Caucasian 1 0.74 (0.50–1.09) NA

OSCC 3 0.94 (0.62–1.43) 16 %/0.30


usual method, based on unadjusted data [7, 8, 13, 14, 35–38], we also extracted the adjusted data and pooledthem for investigating the interactions between gen-etic polymorphisms and environmental risk factors.Interestingly, the unadjusted data showed COX-2rs689466 polymorphism might play a role in in-creased risk while the adjusted data showed a nega-tive association. As we know, smoking and alcoholare the well known risk factors for HNSCC [2, 4].One study by Mittal M et al. [31] adjusted age andgender only, while the other included studies all ad-justed smoking and alcohol. While, there is a relevantmeta-analysis by Zhao F et al. published in 2014 [39].This meta-analysis focused on the association betweenCOX-2 rs20417 polymorphism and digestive systemcancer, including three studies of HNSCC [28, 29, 31]and revealed negative association based on the perform-ance of 2 genetic models (GG + GC vs. GG: OR = 0.66,95 % CI = 0.29, 1.50; C vs. G: OR = 0.95, 95 % CI = 0.56,1.63). Whereas, our meta-analysis performed all recom-mended 5 genetic models, included more studies, andconsidered adjusted data. Furthermore, our meta-analysis investigated 3 polymorphisms at the same timeand only focussed on HNSCC. Different cancers havetheir own histological characteristics and of course theirown predisposing genes. The identical polymorphism inthe same gene, different polymorphisms in the samegene, and identical polymorphism in different genesmight reveal different associations in different cancers.Hence, our meta-analysis was more useful for reference.Also considering this point, we extracted the data forOSCC and LSCC if applicable. The results of all geneticmodels all showed negative association of OSSS, LSCC,and overall population. In addition we considered

genetic background. We stratified the population byethnicity to explore whether different ethnicities havedifferent susceptibility. The results showed all these 3polymorphisms in COX-2 gene regardless of geneticbackground of HNSCC.As we know, COX-2 participated in cell prolifera-

tion and tumour microenvironment and associatedwith many types of cancer. However, our resultsshowed there was non-association of COX-2 andHNSCC. The possible mechanism of the negative re-sult due to the relative small sample size, which isnot enough to detect the small genetic effect. More-over, COX-2 gene polymorphisms were really not as-sociated with HNSCC risk. Third, the compromiseeffect might be existed in the 3 polymorphisms ofCOX-2 or other environmental risk factors, such asgreen tea. Besides, the haplotype analysis was not per-formed because of limited information of includedstudies. However, to explore the true effects and pos-sible mechanism between them remain necessary.Heterogeneity is 1 of the important issues in genetic

association meta-analysis. This limitation also existedin the present meta-analysis, some genetic modelsshowed clear homogeneity while some showed hetero-geneity, either in overall population or subgroup ana-lyses (Tables 3, 4 and 5). The heterogeneity might beoriginated from different genotyping methods, envir-onmental differences, or different lifestyles. However,we could not explore these factors due to the lack ofindividual data. Also, the number of eligible studiesand sample sizes of for each polymorphism was insuf-ficient. Statistical power is influenced by small samplesizes so owing to this limitation, we could not per-form publication bias of any polymorphism. We did

Table 4 Overall and subgroups meta-analysis of COX-2 rs5275 polymorphism and HNSCC risk (Continued)

LSCC 2 1.02 (0.40–2.60) 63 %/0.10

CC + CT vs. TT (unadjusted and adjusted)






Asians (adjusted) 2 1.00 (0.79–1.27) 0 %/0.49



OSCC (unadjusted) 3 1.09 (0.55–2.16) 91 %/



C vs. G (unadjusted and adjusted)

Overall (unadjusted) 5 1.13 (0.62–2.05) 92 %/

not confirm whether relevant publications publishedin languages other than English or Chinese existed,due to lack of right to search and ability to read, assuch we may have missed some eligible studies. Thislimitation was also revealed in the included population.Our meta-analysis only included Asians and Caucasians,hence, our results had no value for other ethnicities. Fi-nally, lacking a relevant recommended tool, we could notassess the methodological quality of included studies anddid not performed subgroup analysis based on high vs.low quality. As such, we did not conduct the meta-regression of methodological quality.

ConclusionIn summary, our meta-analysis based on crude and ad-justed data showed that none of COX-2 rs689466,rs5275, and rs20417 polymorphisms was associated withrisk of HNSCC. Due to limitations of our meta-analysis,such as insufficient sample sizes, our results should betreated with caution. We recommend further high qual-ity studies, with large sample sizes and stratified bysmoking status and alcohol consumption, be conductedto provide high level evidence for clinical implication.

AbbreviationsCI, confidence interval; COX-2, cyclooxygenase-2; HNSCC, Head and necksquamous cell carcinoma; HPV, Human papillomavirus; HWE, Hardy–WeinbergEquilibrium; LSXX, laryngeal squamous cell carcinoma; OR, Odds ratio; OSCC,Oral squamous cell carcinoma; PRISMA, Preferred Reporting Items for SystematicReviews and Meta-Analyses; RevMan, Review Manager

AcknowledgementsWe will thank to the all the authors of the included original studies.

FundingThis research was supported (in part) by the Nature Science Foundation ofHubei Province (2012FFB03902) and the Foundation of Evidence-basedMedicine Nursery Fund of Taihe Hospital (EBM2013002 and EBM20140067),without commercial or not-for-profit sectors. The funders had no role instudy design, data collection and analysis, decision to publish, or preparationof the manuscript. No additional external funding received for this study.

Availability of data and materialMeta-analysis is a secondary analysis which the data are all available from allincluded studies; hence, all the data and material can be found in theincluded original studies.

Authors’ contributionsWDL and XTZ design this manuscript. JSWK and XJW performed systematicliterature search. WH and JGC extracted data and performed statisticalanalysis. WDL and XJW wrote the manuscript. JSWK and XTZ reviewed themanuscript. All authors approved the final manuscript.

Competing interestsThe authors declare that they have no competing interests.

Consent for publicationNot applicable.

Ethics approval and consent to participateNot applicable.

Author details1Department of Stomatology, Taihe Hospital, Institute of Oral andMaxillofacial Surgery, Hubei University of Medicine, Shiyan 442000, China.2Department of Stomatology, Daping Hospital & Research Institute ofSurgery, Third Military Medical University, Chongqing 400042, China.3Chinese Cochrane Center, Chinese Evidence-Based Medicine Center,Western China Hospital, West China Hospital, Sichuan University, Chengdu610041, China. 4Department of Stomatology, Zhuhai People’s Hospital,Zhuhai Hospital Affiliated with Jinan University, Zhuhai 519000, China.5Department of Stomatology, Center for Evidence-Based and TranslationalMedicine, Zhongnan Hospital of Wuhan University, 169 Donghu Road,Wuhan 430071, China.

Received: 3 May 2015 Accepted: 6 July 2016

References1. Zeng XT, Luo W, Huang W, Wang Q, Guo Y, Leng WD. Tooth loss and head

and neck cancer: a meta-analysis of observational studies. PLoS ONE.2013;8(11), e79074.

2. Li Y, Mao Y, Zhang Y, Cai S, Chen G, Ding Y, Guo J, Chen K, Jin M. Alcoholdrinking and upper aerodigestive tract cancer mortality: a systematic reviewand meta-analysis. Oral Oncol. 2014;50(4):269–75.

3. Zeng XT, Deng AP, Li C, Xia LY, Niu YM, Leng WD. Periodontal disease andrisk of head and neck cancer: a meta-analysis of observational studies. PLoSONE. 2013;8(10), e79017.

4. Wyss A, Hashibe M, Chuang SC, Lee YC, Zhang ZF, Yu GP, Winn DM, Wei Q,Talamini R, Szeszenia-Dabrowska N, et al. Cigarette, cigar, and pipe smokingand the risk of head and neck cancers: pooled analysis in the InternationalHead and Neck Cancer Epidemiology Consortium. Am J Epidemiol.2013;178(5):679–90.

Fig. 4 Forest plot for C vs. G model of crude data of rs20417 polymorphism


5. Zeng XT, Leng WD, Zhang C, Liu J, Cao SY, Huang W. Meta-analysis on theassociation between toothbrushing and head and neck cancer. Oral Oncol.2015;51(5):446–51.

6. Abogunrin S, Di Tanna GL, Keeping S, Carroll S, Iheanacho I. Prevalence ofhuman papillomavirus in head and neck cancers in European populations: ameta-analysis. BMC Cancer. 2014;14(1):968.

7. Zeng XT, Luo W, Geng PL, Guo Y, Niu YM, Leng WD. Association betweenthe TP53 codon 72 polymorphism and risk of oral squamous cell carcinomain Asians: a meta-analysis. BMC Cancer. 2014;14:469.

8. Brunotto M, Zarate AM, Bono A, Barra JL, Berra S. Risk genes in head andneck cancer: A systematic review and meta-analysis of last 5 years. OralOncol. 2014;50(3):178–88.

9. Wang Y, Gao X, Wei F, Zhang X, Yu J, Zhao H, Sun Q, Yan F, Yan C, Li H, et al.The hOGG1 Ser326Cys polymorphism contributes to digestive system cancersusceptibility: evidence from 48 case–control studies. Tumour Biol. 2015;36(2):1029-1038.

10. Wu W, Liu L, Yin Z, Guan P, Li X, Zhou B. Association of X-ray repaircross-complementing group 1 Arg194Trp, Arg399Gln and Arg280Hispolymorphisms with head and neck cancer susceptibility: a meta-analysis. PLoS ONE. 2014;9(1), e86798.

11. Zhang E, Cui Z, Xu Z, Duan W, Huang S, Tan X, Yin Z, Sun C, Lu L. Associationbetween polymorphisms in ERCC2 gene and oral cancer risk: evidence from ameta-analysis. BMC Cancer. 2013;13:594.

12. Hu YY, Yuan H, Jiang GB, Chen N, Wen L, Leng WD, Zeng XT, Niu YM.Associations between XPD Asp312Asn polymorphism and risk of headand neck cancer: a meta-analysis based on 7,122 subjects. PLoS ONE.2012;7(4), e35220.

13. Leng WD, He MN, Chen QL, Gong H, Zhang L, Zeng XT. Vascularendothelial growth factor (VEGF) gene polymorphisms and risk of headand neck cancer: a meta-analysis involving 2,444 individuals. Mol BiolRep. 2013;40(10):5987–92.

14. Jiang L, Weng H, Chen MY, Zhang C, Zeng XT. Association betweencyclooxygenase-2 gene polymorphisms and risk of periodontitis: a meta-analysis involving 5653 individuals. Mol Biol Rep. 2014;41(7):4795–801.

15. Fritsche E, Baek SJ, King LM, Zeldin DC, Eling TE, Bell DA. Functionalcharacterization of cyclooxygenase-2 polymorphisms. J Pharmacol Exp Ther.2001;299(2):468–76.

16. Wu H, Wu X, Wan G, Zhang S. Associations between Cox-2 rs20417 andrs5275 polymorphisms and the risk of hepatocellular carcinoma: a metaanalysis. Int J Clin Exp Pathol. 2014;7(10):6898–905.

17. Peng Q, Yang S, Lao X, Tang W, Chen Z, Lai H, Wang J, Sui J, Qin X, Li S.Meta-analysis of the association between COX-2 polymorphisms andrisk of colorectal cancer based on case–control studies. PLoS ONE.2014;9(4), e94790.

18. Dai ZJ, Shao YP, Ma XB, Xu D, Tang W, Kang HF, Lin S, Wang M, Ren HT,Wang XJ. Association of the three common SNPs of cyclooxygenase-2gene (rs20417, rs689466, and rs5275) with the susceptibility of breastcancer: an updated meta-analysis involving 34,590 subjects. Dis Markers.2014;2014:484729.

19. Yang X, Li B, Si T, Liu Y, Guo Z. Association between the 8473 T > Cpolymorphism of PTGS2 and prostate cancer risk: a metaanalysis including24,716 subjects. Onkologie. 2013;36(4):182–6.

20. Yan WF, Sun PC, Nie CF, Wu G. Cyclooxygenase-2 polymorphisms wereassociated with the risk of gastric cancer: evidence from a meta-analysisbased on case–control studies. Tumour Biol. 2013;34(6):3323–30.

21. Yang CY, Meng CL, Liao CL, Wong PY. Regulation of cell growth byselective COX-2 inhibitors in oral carcinoma cell lines. Prostaglandins OtherLipid Mediat. 2003;72(3–4):115–30.

22. Saba NF, Choi M, Muller S, Shin HJ, Tighiouart M, Papadimitrakopoulou VA,El-Naggar AK, Khuri FR, Chen ZG, Shin DM. Role of cyclooxygenase-2 intumor progression and survival of head and neck squamous cell carcinoma.Cancer Prev Res (Phila). 2009;2(9):823–9.

23. Zeng X, Zhang Y, Kwong JS, Zhang C, Li S, Sun F, Niu Y, Du L. Themethodological quality assessment tools for preclinical and clinical studies,systematic review and meta-analysis, and clinical practice guideline: asystematic review. J Evid Based Med. 2015;8(1):2–10.

24. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reportingitems for systematic reviews and meta-analyses: the PRISMA statement. BMJ.2009;339:b2535.

25. Salanti G, Amountza G, Ntzani EE, Ioannidis JP. Hardy-Weinberg equilibriumin genetic association studies: an empirical evaluation of reporting,deviations, and power. Eur J Hum Genet. 2005;13(7):840–8.

26. Huedo-Medina TB, Sanchez-Meca J, Marin-Martinez F, Botella J. Assessingheterogeneity in meta-analysis: Q statistic or I2 index? Psychol Methods.2006;11(2):193–206.

27. Campa D, Hashibe M, Zaridze D, Szeszenia-Dabrowska N, Mates IN,Janout V, Holcatova I, Fabianova E, Gaborieau V, Hung RJ, et al.Association of common polymorphisms in inflammatory genes with riskof developing cancers of the upper aerodigestive tract. Cancer CausesControl. 2007;18(4):449–55.

28. Chiang SL, Chen PH, Lee CH, Ko AM, Lee KW, Lin YC, Ho PS, Tu HP, Wu DC,Shieh TY, et al. Up-regulation of inflammatory signalings by areca nutextract and role of cyclooxygenase-2 -1195G > a polymorphism reveal riskof oral cancer. Cancer Res. 2008;68(20):8489–98.

29. Lin YC, Huang HI, Wang LH, Tsai CC, Lung O, Dai CY, Yu ML, Ho CK,Chen CH. Polymorphisms of COX-2 -765G > C and p53 codon 72 andrisks of oral squamous cell carcinoma in a Taiwan population. OralOncol. 2008;44(8):798–804.

30. Peters WH, Lacko M, Te Morsche RH, Voogd AC, Oude Ophuis MB, Manni JJ.COX-2 polymorphisms and the risk for head and neck cancer in whitepatients. Head Neck. 2009;31(7):938–43.

31. Mittal M, Kapoor V, Mohanti BK, Das SN. Functional variants of COX-2 andrisk of tobacco-related oral squamous cell carcinoma in high-risk AsianIndians. Oral Oncol. 2010;46(8):622–6.

32. Lakshmi A, Muralidhar S, Kalyan Kumar C, Pavan Kumar A, KalyanaChakravarthy P, Anjaneyulu V, Kaiser J. Cyclooxygenase-2-765G > Cfunctional promoter polymorphism and its association with oralsquamous cell carcinoma. J Investig Clin Dent. 2012;3(3):182–8.

33. Chang JS, Lo HI, Wong TY, Huang CC, Lee WT, Tsai ST, Chen KC, Yen CJ,Wu YH, Hsueh WT, et al. Investigating the association between oralhygiene and head and neck cancer. Oral Oncol. 2013;49(10):1010–7.

34. Niu Y, Yuan H, Shen M, Li H, Hu Y, Chen N. Association betweencyclooxygenase-2 gene polymorphisms and head and neck squamous cellcarcinoma risk. J Craniofac Surg. 2014;25(2):333–7.

35. Zeng XT, Liu DY, Kwong JS, Leng WD, Xia LY, Mao M. Meta-Analysis ofAssociation Between Interleukin-1beta C-511 T Polymorphism and ChronicPeriodontitis Susceptibility. J Periodontol. 2015;86(6):812–9.

36. Mandal RK, Yadav SS, Panda AK, Khattri S. Vascular Endothelial GrowthFactor 936 C > T Polymorphism Increased Oral Cancer Risk: Evidence from aMeta-Analysis. Genet Test Mol Biomarkers. 2013.

37. Xia LY, Zeng XT, Li C, Leng WD, Fan MW. Association between p53Arg72Pro polymorphism and the risk of human papillomavirus-related headand neck squamous cell carcinoma: a meta-analysis. Asian Pac J CancerPrev. 2013;14(10):6127–30.

38. Yan Y, Weng H, Shen ZH, Wu L, Zeng XT. Association between interleukin–4gene–590 c/t,–33 c/t, and 70–base-pair polymorphisms and periodontitissusceptibility: a meta-analysis. J Periodontol. 2014;85(11):e354–62.

39. Zhao F, Cao Y, Zhu H, Huang M, Yi C, Huang Y. The–765G > CPolymorphism in the Cyclooxygenase–2 Gene and Digestive SystemCancer: a Meta-analysis. Asian Pac J Cancer Prev. 2014;15(19):8301–10.

• We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal• We provide round the clock customer support • Convenient online submission• Thorough peer review• Inclusion in PubMed and all major indexing services • Maximum visibility for your research

Submit your manuscript atwww.biomedcentral.com/submit

Submit your next manuscript to BioMed Central and we will help you at every step:


AbstractBackgroundMethodsResultsConclusion

BackgroundMethodsEligibility criteriaSearch strategyData extractionStatistical analysis

ResultsStudy identification and characteristicsCOX-2 rs689466 polymorphism and HNSCC riskCOX-2 rs5275 polymorphism and HNSCC riskCOX-2 rs20417 polymorphism and HNSCC riskPublication bias

DiscussionConclusionAbbreviationsAcknowledgementsFundingAvailability of data and materialAuthors’ contributionsCompeting interestsConsent for publicationEthics approval and consent to participateAuthor detailsReferences

Documents

COX-2 rs689466, rs5275, and rs20417 polymorphisms and …Wei-Dong Leng1†, Xiu-Jie Wen2†, Joey S. W. Kwong3†, Wei Huang4, Jian-Gang Chen5 and Xian-Tao Zeng1,5* Abstract Background: